Public concern about Autism Spectrum Disorders has been fueled by reports of increased prevalence in school age and preschool children. While the onset of ASD is within the first36 months of life, prevalence during the first 3 years of life is not established. This proposal aims to develop and pilot test methods for ASD surveillance in early childhood, in preparation for a subsequent prevalence study of a single year birth cohort. Emerging data indicate that autism can be identified in children under age 3. However, whether it is evident in the 1st, 2nd, or 3rd year of life can vary, especially if the child experiences a developmental regression. The current application presents a design for a population-based prevalence study that would follow children from ages 18-36 months for a 2 year period, with repeated screenings throughout to capture children as their autism manifests. Specific aims for the current proposal are to pilot test this screening approach, and to establish the relationships and data linkages necessary to conduct the full prevalence study. In addition, we will complete the record review methods used in other CDC prevalence studies and compare case rates between direct screening and record review only to determine whether record review alone is sufficient in this age range. This study responds to several of the priorities outlined by the lnteragency Autism Coordinating Committee, including a call for intensive, community-based prevalence studies with clinical evaluations that can be repeated over time to determine changes in prevalence, methods to develop community-based screening of very young children, and testing of screening tools. This application builds on a foundation of Utah research on autism that includes: ADDM network Study Year 2002; an established relationship between the University and the Health Department; widespread community support for autism research; and a strong University research environment. [unreadable] [unreadable] [unreadable] [unreadable]